US 7007278 B2
Interactive legacy applications can be run from a network, such as the Internet, without requiring any code changes in the application. Typically, legacy applications are critical to a business, are self-contained on the computer, have mixed business and user interface logic, and were written before distributed computing emerged. Separating business logic from user interface logic as required by web application architectures is not practicable in the case of legacy applications. A client has a network user agent which can access a network server connected to the computer. When an application is invoked from the network user agent, a runtime data redirector intercepts the application's raw data and sends the data to the network server which then serves the data across the network to the network user agent. Input data from the user entered through the network user agent are sent back to the application via the same runtime intercept.
1. A method for accessing a legacy computer application over an Internet, the method comprising:
converting a proprietary screen definition to a converted user interface page in a pervasive computer Internet user agent format, the converting step further comprising:
parsing the proprietary screen definition associated with the legacy computer application, wherein the proprietary screen definition is stored separately from the legacy computer application;
mechanically mapping user interface elements of the proprietary screen definition to an extensible mark-up language (XML) based language that supports user interface elements;
customizing the converted user interface page using a customize template that identifies user interface design patterns that reflect a presentation style of a user; and
validating input fields of the converted user interface page;
providing access to the legacy computer application responsive to a request from a client that is separate from a computer on which the legacy computer application resides, the client and the computer being connected by a server, the providing step further comprising:
redirecting a raw output of the legacy computer application to a network publishing component;
reformatting the raw output for publishing, wherein the reformatted raw output is stored in JavaBean data objects that are populated by a servlet which has received data to publish;
updating dynamically the convened user interface pages using the reformatted raw output;
sending the updated converted user interface pages to the client; and
reformatting input data from the client using the network publishing component to a format of the legacy computer application; and
making the reformatted raw output in the JavaBean data objects available for access, use and manipulation by another network based application;
wherein the legacy computer application is not restructured and no code changes are made to programs of the legacy computer application,
wherein the legacy computer application is non-modular,
wherein the legacy computer application scamlessly interweaves with the network based application, and
wherein a state of the legacy computer application is automatically preserved in a native environment after a transaction on the Internet is completed and a connection to the Internet is broken.
This application relates generally to the field of computer software applications and more particularly relates to accessing legacy application programs over a computer network, such as the Internet.
Like everything else, computer use has changed over the years. In the early days, large mainframe computers dominated the industry until the advent of the personal standalone computer. Now many businesses and homes have at least one personal standalone computer, a PC. A new paradigm of computing, however, has emerged: network-centric computing or distributed computing in which at least two, but more likely many more computers, called clients and servers, are interconnected through a network wherein the software applications used by a client resides on a server. Thus, a server may provide the application or client program or databases used by an end user or by a number of other servers and clients over a network.
In many instances and increasingly more so, the network connecting clients and servers is the Internet. The Internet refers to a collection of interconnected computer networks that use the Internet protocol, e.g., TCP/IP, UDP, etc. Software applications are written in a programming language independent of the operating system of either the server or the client to access and use a myriad of applications on the Internet. Languages that describe data over the Internet, moreover, include variations of extensible mark-up languages (XML), such as hypertext mark-up language (HTML), wireless mark-up language (WML), etc. The world wide web refers to a software management scheme which accesses the Internet with various user agents using hypertext links.
Distributed computing has fundamentally changed the methodology of software developers in writing code for these applications. Rather than writing one massive self-contained application having thousands or millions of lines of code, software developers select and weave together smaller portions of code, called components, each associated with a particular function that can be reused across multiple applications, all on the world wide web. Components are standard software that can be pulled off a server and incorporated into new applications by software developers. For example, a calendar component may be used in several applications such as a scheduling application, a presentation application, and/or a data base application to calculate employee's vacation and pay, etc. Current industry architectures for Internet and world wide web applications, moreover, require that business logic be separated from presentation logic such that the business logic is solely responsible for providing business data while the presentation logic is solely responsible for driving presentation or the user interface.
The massive self-contained applications having thousands or millions of lines of code, however, may still be very important to businesses and organizations. Indeed, some self-contained programs may contain a business's entire inventory/client list/database or all of the above. The term “legacy applications” refers to old applications which remain in operation with a business or organization but were developed before the advent of distributed computing and structured programming and its techniques of modularity, coupling, cohesion, etc. Legacy applications written as one large self-contained program do not accommodate change easily. These legacy applications, moreover, are often so critical to the business that these businesses face difficult decisions about the future of such applications and the ability to exploit new network and Internet technologies for remote and distributed access to these critical applications. Interactive legacy applications that have mixed business and presentation logic must be restructured and/or reengineered to be enabled as web applications. Separating the business logic from the user interface logic, however, is not always practicable to deploy applications onto the web with minimal intervention and testing.
Other solutions to implement legacy applications on the Internet and world wide web may use a screen-scraping technique which converts legacy display data streams at runtime to a web-based language, such as HTML, or other language. The HTML converted screens are then displayed on a browser. Converting legacy display data stream at runtime, however, is slow, compromises performance of the application, and does not give the user the capability of integrating or bridging legacy applications seamlessly with new web or other network applications. Nor does mere conversion of legacy data provide the capability to enhance the effectiveness of the user interface of the application when the application runs on the Internet.
There is thus a need in the industry to access legacy applications from the Internet, the world wide web, or other computer communication networks without having to restructure the application or make code changes to the programs.
There is a further need in the industry to seamlessly interweave legacy applications with other network applications.
These needs and others that will become apparent to one skilled in the art are satisfied by a method to execute a computer application installed on a computer, comprising the steps of creating a servlet instance in a server connected to the computer on a first network; running the application on the computer to generate dynamic data; intercepting and redirecting the dynamic data to a network publishing component on the computer; transmitting dynamic data from the network publishing component to the servlet instance; and creating data objects and populating the data objects with the dynamic data in the server. The method may further comprise requesting the application from a client connected to the server over a second network; updating at least one network page with the dynamic data; and transmitting the updated network pages to the client.
The first network may be the Internet. The first network may also be selected from the group consisting of: an internal network, an Intranet, a LAN, a WAN, an internal bus, a wireless network. Similarly the second network may be the Internet; or may be selected from the group consisting of: an internal network, an Intranet, a LAN, a WAN, an internal bus, a wireless network. The computer may contain the network server.
The method may further comprise converting the display files of the application to network pages capable of displaying dynamic data. The network pages may be based on a XML language, such as HTML and/or WML. The network pages may be JavaServerPages. The network pages, moreover, may be stored on the server.
The method of the invention may further comprise creating an I/O buffer for the dynamic data in the computer. The method may also comprise the steps of creating a first endpoint connection between the servlet instance and the network publishing component; the endpoint connection may be a socket, or a data queue object, or a message queue.
It is also envisioned that the invention comprises a program product for use in a computer network for executing an application stored on a computer from a client, the computer program product comprising a signal-bearing medium carrying thereon: an application invoker to start and run an application in its native environment on the computer from a client; a plurality of network user interface pages to display the application's input/output data on the client; a data redirector to redirect the application's input/output data to network user-interface pages; a plurality of data objects corresponding to the network user interface pages to receive the application's input/output data; a servlet instance to dynamically update the network user interface pages with the application's input/output data; and a network user agent to display the updated network user-interface pages on the client. The program product may further comprise a screen definition converter to convert the input/output screen definitions of the application to the network user-interface pages.
The invention may also be considered a computer system for executing an application, comprising: a central processing unit; a main memory connected to the central processing unit with a communication bus; a data storage unit connected to a data storage interface which is connected to the communication bus; at least one input/output device connected to the communication bus and connected to a network interface to an external computer network, an application stored in the main memory and capable of executing on the central processing unit; a network publishing component; a data redirector to redirect the application's dynamic data to the network publishing component; and an I/O buffer to store the redirected dynamic data.
Yet, in another embodiment, the invention may be considered a network server for accessing an application stored and executing on a computer, comprising: a central processing unit; a network interface to connect to at least one client over a network; a servlet instance to receive a request from the at least one client to access the application and transmit the request to the computer; a server endpoint connection for transmitting and receiving real-time data to and from the computer on which the application is executing; and a plurality of data objects to be populated with the real-time data wherein the servlet receives the real-time data from the application and populates the data objects with the real-time data. The network server may further comprise a plurality of network display pages, each of the network display pages unique to each input/output screen definition of the application wherein the servlet updates the network display pages for transmission to the client over the network.
The invention may also be viewed as a method for executing a computer application installed on a computer, comprising the steps of: converting a plurality of display files of the application to a plurality of XML-based network pages capable of displaying the application's dynamic data; creating a servlet instance in a server connected to the computer on a network; requesting the application from a client connected to a server over the Internet; running the application on the computer in its native environment; creating an I/O buffer in the computer for the application's dynamic data; creating an endpoint connection between the servlet instance and a network publishing component on the computer; transmitting the dynamic data back and forth from the client to the application through the servlet instance; creating data objects and populating the data objects with the dynamic data; updating at least one network page using the data objects; transmitting the updated network pages to the client over the Internet; and transmitting network pages having responsive data from the client to server for transmission as input data to the application.
Yet, another aspect of the invention is a method of interacting with a computer application, comprising: executing a legacy computer application in its native environment; redirecting I/O requests from and responses to the legacy computer application from a client over the Internet without introducing changes to the code of the legacy computer application.
Referring to the Drawing, wherein like numbers denote like parts throughout the several views,
Terminal interface 103 may support the attachment of single or multiple terminals and may be implemented as one or multiple electronic circuit cards or other units. Data storage 104 preferably comprises one or more rotating magnetic hard disk drive units, although other types of data storage could be used. Network interface 105 provides a physical connection for transmission of data to and from a network. In the preferred embodiment the network is the Internet but the network could also be any smaller self-contained network such as an Intranet, a WAN, a LAN, or other internal or external network using, e.g., telephone transmissions lines, satellites, fiber optics, T1 lines, etc. and any various available technologies. Communications bus 110 supports transfer of data, commands and other information between different devices; while shown in simplified form as a single bus, it is typically structured as multiple buses; and may be arranged in a hierarchical form.
The computer system shown in
Computer system 100 and its components are shown and described in
With reference to
As will be described in detail below, aspects of the preferred embodiment pertain to specific method steps implementable on a computer 100 or a network server 200. In an alternative embodiment, the invention may be implemented as a computer program-product for use with either or both a network server 200 and a client 300. The programs defining the functions of the preferred embodiment can be delivered to the computer 100 and/or to the network server 200 via a variety of signal-bearing media, which include, but are not limited to: (a) information permanently stored on non-writable storage media, e.g., read only memory devices within either computer such as CD-ROM disks readable by CD-ROM drive 214; (b) alterable information stored on writeable storage media, e.g., floppy disks within diskette drive or hard-disk drive, such as shown as 214 in
This invention provides an environment that shields interactive legacy applications from restructuring or reengineering the applications' code by allowing such legacy applications to continue running, as-is, in their native environment while simultaneously allowing a variety of clients to access the legacy applications using a web browser or other network technology. The invention also provides the interactive legacy applications with the ability to generate the required business data and drive separate presentation artifacts as required by industry standard world wide web application architectures. What is interesting about the invention is the interaction between the “stateful” legacy applications and the “stateless” Internet transactions. Interactive legacy applications are stateful in that the application continues to execute until it comes to an I/O instruction wherein execution is suspended and the application's state is preserved in its native environment until the I/O action is completed. Internet transactions, on the other hand, use stateless protocols of, e.g., HTTP, wherein after a transaction on the Internet is completed, the connection to the Internet is broken and the state of the transaction is lost unless additional processing occurs to save that state on the network server. This invention provides a mechanism and an environment whereby the state of the legacy application is automatically preserved in its native environment while it converses with the stateless web environment. The automatic preservation of the application's state is inherent in the architecture of the invention which provides the necessary continuity required by interactive legacy applications.
With reference to
The converted user interface pages 520 are dynamically updated at step 530 using existing web page serving technologies on the network server 200 prior to sending the page to the network user agent 570 on the client 300. The application's raw output data are redirected by the data manager 432 of the application runtime operating system 430 at step 540 so that the outbound data for display are not converted to traditional proprietary display data streams as described in
In a particular embodiment of the Internet and world wide web as the network, JAVA allows JavaServer Pages as the network pages to dynamically insert application data into HTML or XML pages before the network pages are served to a network user agent, a process referred to as dynamically generated web content. The data to be read by the JavaServer Pages are stored in data objects, i.e., JavaBeans, that are populated by a servlet which has received data to publish. A very important feature of the invention is that once the legacy application data has been stored in these data objects in the network server, it becomes available for any purpose, including access from another computer connected to the server, for manipulation, for transmission . . . the uses for this legacy dynamic data on the web server are limitless. The data field definitions of the JavaBeans have a one-to-one mapping with the variable data fields of HTML or XML pages housed by the JavaServer Pages and are generated during the conversion process as Java class definitions because the definition of the proprietary application screens contain the I/O fields of the screens. Such mapping is known in the art, and there are other products such as Active Server Pages that are capable of providing dynamic content from a server to a client. In any embodiment, the servlet instance instantiates and populates these data objects with the redirected dynamic data, usually I/O data, of the legacy application. Preferably, therefore, the data objects have the necessary methods to format the data for display and for converting the data from the application's proprietary code to the network codepage of the appropriate language.
By way of example only,
The control flow of the process in accordance with principles of the invention proceeds as follows: a user on a client 300 starts the legacy application 122 from the network user agent 570, preferably a world wide web browser, by first requesting at step 612 an invocation page from the servlet instance 610 on a network server 200. In step 614, the client 300 which may be another computer or a person completes and submits a set of entry fields for the name and parameters of the application to invoke. The servlet instance 610, upon receiving the XML or HTTP request for invocation, at step 616 creates a socket 626 a and spawns a secondary thread 618 to the computer 100 launch the application process that invokes the legacy application. The socket 626 a then waits for a connection as in step 642 a.
The secondary thread 618 passes the socket port number and program information to the application invoker 660 which creates a socket 626 b and makes a connection to the socket in the network server process as in step 620. After a connection is established in step 622 between the two sockets 626 a and 626 b, the socket 626 b in the network server process waits for data from the legacy program. The application invoker 660 writes the descriptor of the application-side socket 626 b to an environment variable, activates data redirection through an application programming interface and invokes the legacy application 122, as shown in step 624. When the legacy application 122 reaches an I/O instruction, output data is sent as in step 632 to the application runtime component 430 of the computer's operating system which calls the Write_Data method as in 640 a to redirect data to the application-side socket 626 b. Data is read by the socket 626 a in the network server process as in step 642 b and the servlet instance 610 creates a data object in step 650 to store the data. In step 652, the servlet instance 610 calls the network page, e.g., the JavaServer Page, associated with this output record. The network page is populated with data from the data object as in step 652 and then served to the network user agent 570 as, e.g., an HTTP response, in step 660.
Subsequent interactions between the client interface on the network user agent 570 and the application 122 flows through the socket connections 626 a and 626 b. When the client submits a page, the servlet instance 610 on the network server restores session data from a previously saved session object which has a reference to a socket or a data queue object or a message queue object. The input data is then forwarded to socket or queue 626 a as in step 642 c, to the other application socket or queue 626 b and I/O buffers if any and to the application runtime component 430, and eventually to the legacy program 122 that was waiting on a Read_Data method 640 b. The legacy application 122 then continues its execution in its native environment on the computer 100 until the next I/O exchange. When the application ends, control returns to the application invoker 660 which closes the endpoint connection as in step 620.
Thus, the invention allows the user to deploy entire business-critical legacy applications to the Internet to take advantage of state-of-the art technology without having to restructure or make code changes to the programs. The legacy application, moreover, executes normally. Legacy application data becomes available across the network for any purpose because it is stored in data objects on the network server. The invention, moreover, allows the user to modify the network pages, such as JavaServer Pages, to create links within the network server to new web based applications. This capability allow interactions between legacy and other network based applications, including merging legacy application data with data from other network applications in the network server. Computer and client systems that allow remote invocation of programs are able to use the architecture described above to run the web server and the legacy applications in separate computer systems. The invention, moreover, allows the user to customize the JavaServer Pages according to the company's rules and guidelines for application presentation. The solution presented herein further allows the I/O data to be converted to other formats that would suit display types other than a browser. One powerful application is the conversion of the I/O data and the display records to a generic XML based user interface. The XML-user interface can then be fed into different types of user interface renderers. These renderers can be Java-Swing based, Voice-Based, PDA Based, etc. The invention, therefore, opens access to existing legacy application from multiple types of future devices.
While various embodiments of the present invention have been described above, it should be understood that they have been presented by way of example and not limitation and that variations are possible. Thus, the breadth and scope of the present invention should not be limited by any of the above-described exemplary embodiments, but should be defined only in accordance with the following claims and their equivalents.